Optimizing the number of equivalent iterations of 3D OSEM in SPECT reconstruction of I-131 focal activities

Koral, Kenneth F.; Kritzmaan, James N.; Rogers, Virginia; Ackermann, Robert; Fessler, Jeffrey A.

doi:10.1016/j.nima.2007.04.070

Cited by 13 publications

(11 citation statements)

References 5 publications

Supporting

Mentioning

Contrasting

Order By: Relevance

“…The large distortions near the target edges with OSEM become more severe as iterations proceed and have been reported previously in EM reconstructions of objects with sharp edges (Snyder et al 1987, Tsui et al 1994, Koral et al 2007a, Shcherbinin and Celler 2008). PL-CT images and profiles did not display the significant edge overshoots or the blurring across region boundaries evident in the methods where CT information was not used in the regularizer.…”

Section: Discussionsupporting

confidence: 73%

“…Indiscriminate smoothing including across region boundaries degrades recovery of activity. However, images produced by iterative reconstruction without filtering have been observed to have edge artifacts that worsen as the iterations proceed (Snyder et al 1987, Tsui et al 1994, Koral et al 2007a). These artifacts are caused by the unstable ‘deconvolution’ of the collimator–detector response when one tries to compensate for that response to improve activity recovery.…”

Section: Introductionmentioning

confidence: 99%

See 1 more Smart Citation

Regularized reconstruction in quantitative SPECT using CT side information from hybrid imaging

2010

View full text Add to dashboard Cite

A penalized-likelihood (PL) SPECT reconstruction method using a modified regularizer that accounts for anatomical boundary side information was implemented to achieve accurate estimates of both the total target activity and the activity distribution within targets. In both simulations and experimental I-131 phantom studies, reconstructions from (1) penalized likelihood employing CTside information-based regularization (PL-CT), (2) penalized likelihood with edge preserving regularization (no CT) and (3) penalized likelihood with conventional spatially invariant quadratic regularization (no CT) were compared with (4) ordered subset expectation maximization (OSEM), which is the iterative algorithm conventionally used in clinics for quantitative SPECT. Evaluations included phantom studies with perfect and imperfect side information and studies with uniform and non-uniform activity distributions in the target. For targets with uniform activity, the PL-CT images and profiles were closest to the 'truth', avoided the edge offshoots evident with OSEM and minimized the blurring across boundaries evident with regularization without CT information. Apart from visual comparison, reconstruction accuracy was evaluated using the bias and standard deviation (STD) of the total target activity estimate and the root mean square error (RMSE) of the activity distribution within the target. PL-CT reconstruction reduced both bias and RMSE compared with regularization without side information. When compared with unregularized OSEM, PL-CT reduced RMSE and STD while bias was comparable. For targets with non-uniform activity, these improvements with PL-CT were observed only when the change in activity was matched by a change in the anatomical image and the corresponding inner boundary was also used to control the regularization. In summary, the present work demonstrates the potential of using CT side information to obtain improved estimates of the activity distribution in targets without sacrificing the accuracy of total target activity estimation. The method is best suited for data acquired on hybrid systems where SPECT-CT misregistration is minimized. To demonstrate clinical application, the PL reconstruction with CTbased regularization was applied to data from a patient who underwent SPECT/CT imaging for tumor dosimetry following I-131 radioimmunotherapy.

show abstract

Section: Discussionsupporting

confidence: 73%

Section: Introductionmentioning

confidence: 99%

Regularized reconstruction in quantitative SPECT using CT side information from hybrid imaging

2010

View full text Add to dashboard Cite

show abstract

“…From the data obtained by the previous equations, we computed the root mean square error (RMSE) for the activity of each sphere as [28]:…”

Section: Discussionmentioning

confidence: 99%

“…Several investigations have been reported on the performance of such algorithm on radionuclides frequently used in nuclear medicine, for patient imaging as well as patient specific dosimetry. Some of the more pertinent publications include improvements in correction techniques for quantification in SPECT [4,[17][18][19][20], evaluations of the accuracy of quantitative image reconstruction techniques [21][22][23][24][25] and comparisons of resolution between SPECT/CT systems using machine-specific reconstruction algorithms [26][27][28]. In these studies, the OSEM reconstruction parameters (i.e.…”

Section: Introductionmentioning

confidence: 99%

Impact of a commercial 3D OSEM reconstruction algorithm on the 177Lu activity quantification of SPECT/CT imaging in a Molecular Radiotherapy trial

Grassi¹,

Fioroni²,

Mezzenga³

et al. 2017

Radiol Diagn Imaging

View full text Add to dashboard Cite

The purpose of this study was to investigate the influence of the Ordered Subsets Expectation Maximization (OSEM) reconstruction updates implemented in the 177 Lu SPECT/CT imaging processing in molecular radiotherapy. A NEMA IEC Body Phantom TM was used to quantify activity in refillable spheres of five different sizes. Images were obtained with a hybrid dual-head SPECT-CT imaging system (Symbia T2, Siemens Medical System, Germany) with a clinical acquisition protocol, and reconstructed using a commercial 3D OSEM algorithm (Flash 3D). In the reconstruction process, different values of iterations and subsets were considered, along with a 3D Gaussian post-reconstruction filter and scatter and attenuation correction.Activity recovery coefficients were derived from the ratio between total reconstructed counts and the true activity for each sphere at each OSEM update. Recovery coefficients, and average fractional error (i.e. the weighted Root Mean Squared Error) were evaluated.At the same time, also 177 Lu spatial resolution and dead time were investigated, as matter of discussion about activity recovery coefficients.Results for spheres ≤ 5.5 ml in volume were significantly affected by the partial volume effect, causing a great bias in activity estimation for the smallest spheres. Their weighted fractional error was OSEM update dependent, ranging between 85% to 79% and 60% to 50% for the two smallest spheres, referring to values of 8 subsets-8 iterations and 16 subsets-10 iterations for the two extremes, respectively. No dead time was detected.The choice of iterations and subsets is dependent on the object size to investigate and on the desired image quality. Anyway, using a fixed number of iterations and subsets is correct for objects with volumes ≥ 5.5 ml, reaching the total count convergence in the reconstructed volumes, but the use of correction factors for compensating the partial volume effect is needed. For objects with volumes ≤ 5.5 ml the quantification becomes challenging.

show abstract

“…An experimental demonstration of the overshoot artifact comes from SPECT acquisitions for a cylindrical phantom with an elliptical cross section that contained seven spheres [3]. The volumes of the spheres ranged from 209 to 4.2cc; they were well separated from each other and arranged at the center and six points of a hexagon as viewed from the top of the phantom.…”

Section: Experimental Demonstration Of Overshootmentioning

confidence: 99%

Nature of Ringing Artifacts in Spect Reconstruction and Their Reduction by the Use of Side Information

Koral

2007

2007 4th IEEE International Symposium on Biomedical Imaging: From Nano to Macro

View full text Add to dashboard Cite

ML-EM iterative reconstruction algorithms tend towards ringing artifacts, and high noise, after a large number of iterations just as filtered backprojection algorithms tend toward streak artifacts. OS-EM is a type of ML-EM algorithm and an edge overshoot has been observed experimentally for uniform-activity spheres of large size but not smaller size. Herein, we shed light on such artifacts by showing that they are a function of the object size relative to the FWHM of the collimator-detector response function. That is, for a given number of iterations, an over-or undershoot can occur at a given spatial location or not depending on the value of the ratio. We further show that a new, regularized, penalized-likelihood algorithm based on ordered subsets and employing side information stabilizes the iterative procedure against the development of artifacts.

show abstract

Optimizing the number of equivalent iterations of 3D OSEM in SPECT reconstruction of I-131 focal activities

Cited by 13 publications

References 5 publications

Regularized reconstruction in quantitative SPECT using CT side information from hybrid imaging

Regularized reconstruction in quantitative SPECT using CT side information from hybrid imaging

Impact of a commercial 3D OSEM reconstruction algorithm on the 177Lu activity quantification of SPECT/CT imaging in a Molecular Radiotherapy trial

Nature of Ringing Artifacts in Spect Reconstruction and Their Reduction by the Use of Side Information

Contact Info

Product

Resources

About